Essays on Infrastructure and Urban Economics
This study examines the effects of infrastructure improvements on various outcome measures of economic performance. I focus on three different examples: (1) the opening of the aviation system in the United States, (2) the effects of improving or labeling certain airports as "hub" airports, and (3) improvements in decades-old public school buildings for energy efficiency and sustainability. Together, each case provides substantial evidence that infrastructure is an important input in the functioning and/or performance of economic activity.
The first chapter considers the effects of small and mid-size commercial airports on their local economies over the post World War II period, specifically 1950-2010. To estimate these effects, I use a detailed, novel dataset of Census Based Statistical Area (CBSA) level employment outcomes, geographic, transportation, and city characteristics, along with previously unexploited historical aviation data. Using an instrumental variables approach, one-to-one Mahalanobis distance matching with caliper and pooled synthetic controls, I show that airports have had substantial effects on CBSA population and employment over time. The larger effect on tradable industry employment implies that the overall employment and population effects may result from direct effects on tradable sector industry productivity, perhaps by facilitating information flows. Effects vary by initial city size and region, and are generally robust to the choice of instruments and/or estimator.
The second chapter considers the marginal effect of having a hub designation by an airline on its cities economic fortunes relative to cities that have airports, but not hub airports. Using panel regression methods and event study techniques, I find that while hub airports do not significantly affect city employment levels, hubs do contribute 1-2 percent of personal income to their respective cities, as well as establishment growth of 1-2 percent. I find the effects of hubs on employment to be most salient in the air transportation and hotel industries; however, the same is not necessarily true for other sectors where tourism might affect employment. This implies that the effects of hub airports, in most cases, operate through their ability to facilitate business travel, as hubs increase non-stop market access by at least 15 percent.
The third chapter considers an infrastructure improvement of a different type: improvements to public school buildings to increase sustainability and reduce energy costs. Said improvements, such as improving ventilation systems, temperature control, and adding more sunlight, are thought to enhance student learning outcomes. To test this, two panel data sets are created: a nationwide panel with school districts as the unit of observation, and a California panel with schools themselves as the unit of analysis. Panel data methods including fixed-effects regression and event study techniques exploit differences in conversion timing to examine the schools' effect on dropout rates, test scores, and school quality indices. Nationwide, I find evidence that energy cost reductions may not be the primary factor driving adoption of green schools. Additionally, considering the evidence from California, it appears that in general energy efficient school buildings have a negligible effect on academic performance, even after looking at a variety of measures, suggesting that sustainable buildings are no panacea for improving school performance.
Taken together, this study demonstrates that infrastructure can affect economic performance. Larger interventions will have a larger effect, while more marginal interventions may have smaller, or even negligible, effects.